{"id":{"repo_id":"vt","oai_identifier":"oai:vtechworks.lib.vt.edu:10919/31484"},"canonical_url":"https://search.dev.ndltd.org/etd/vt/oai:vtechworks.lib.vt.edu:10919/31484","repository":{"repo_id":"vt","name":"Virginia Tech","base_url":"https://vtechworks.lib.vt.edu/oai/request"},"display":{"title":"MPIOR: A Framework to Analyze File System Performance of MPI Applications","abstract":"MPI I/O replay (MPIOR) is an I/O performance modeling and prediction tool used to trace and replay a parallel application to determine application performance under a new I/O sub system. The trace collector deduces synchronization inter-dependencies between nodes and I/O demands placed by each node on the storage subsystem. It uses a novel runtime graph traversal technique to filter and log only those MPI calls that affect I/O, thus substantially reducing both the number of runs and the size of the trace file. Unlike other such tools, MPIOR collects a valid trace in a single run and it does not rely on node sampling or I/O sampling. MPIOR's post processing engine analyzes the trace files and sets up the re-player. Due to minimal overhead for trace collection, MPIOR can be used during production runs rather than just as a debugging tool. The re-player mimics the behavior of the application across a variety of storage systems by mapping multiple processes to multiple threads running on a single node. We show average replay error for parallel applications is below 30%.","abstract_html":"MPI I/O replay (MPIOR) is an I/O performance modeling and prediction tool used to trace and replay a parallel application to determine application performance under a new I/O sub system. The trace collector deduces synchronization inter-dependencies between nodes and I/O demands placed by each node on the storage subsystem. It uses a novel runtime graph traversal technique to filter and log only those MPI calls that affect I/O, thus substantially reducing both the number of runs and the size of the trace file. Unlike other such tools, MPIOR collects a valid trace in a single run and it does not rely on node sampling or I/O sampling. MPIOR's post processing engine analyzes the trace files and sets up the re-player. Due to minimal overhead for trace collection, MPIOR can be used during production runs rather than just as a debugging tool. The re-player mimics the behavior of the application across a variety of storage systems by mapping multiple processes to multiple threads running on a single node. We show average replay error for parallel applications is below 30%.","abstract_has_math":false,"creators":["Banerjee, Shankha"],"institution":"Virginia Tech","degree_name":"Master of Science","degree_level":"masters","degree_discipline":"Computer Science and Applications","degree_department":"Computer Science and Applications","school":null,"contributors":[],"advisors":[],"committee_chairs":["Varadarajan, Srinidhi"],"committee_members":["Tilevich, Eli","Ribbens, Calvin J."],"year":2012,"date_issued":"2012-02-17","date_published":"2012-02-17","updated_at":"2026-07-22T22:18:52Z","subjects":["I/O","replay","MPI","trace"],"languages":[],"rights":["In Copyright"],"rights_urls":["http://rightsstatements.org/vocab/InC/1.0/"],"identifier_entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["etd-03162012-020243"],"render_values":[{"text":"etd-03162012-020243","href":null,"code":true}]}]},"links":{"outbound_url":"http://hdl.handle.net/10919/31484","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.committeechair","label":"Committee Chair","values":["Varadarajan, Srinidhi"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Tilevich, Eli","Ribbens, Calvin J."]},{"key":"dc:contributor.department","label":"Department","values":["Computer Science and Applications"]},{"key":"dc:creator","label":"Author","values":["Banerjee, Shankha"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2014-03-14T20:32:40Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2014-03-14T20:32:40Z","2012-04-11"]},{"key":"dc:date.issued","label":"Date","values":["2012-02-17"]},{"key":"dc:publisher","label":"Institution","values":["Virginia Tech"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science and Applications"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Virginia Polytechnic Institute and State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["I/O","replay","MPI","trace"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://rightsstatements.org/vocab/InC/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["etd-03162012-020243"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/10919/31484"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["MPI I/O replay (MPIOR) is an I/O performance modeling and prediction tool used to trace and replay a parallel application to determine application performance under a new I/O sub system. The trace collector deduces synchronization inter-dependencies between nodes and I/O demands placed by each node on the storage subsystem. It uses a novel runtime graph traversal technique to filter and log only those MPI calls that affect I/O, thus substantially reducing both the number of runs and the size of the trace file. Unlike other such tools, MPIOR collects a valid trace in a single run and it does not rely on node sampling or I/O sampling. MPIOR's post processing engine analyzes the trace files and sets up the re-player. Due to minimal overhead for trace collection, MPIOR can be used during production runs rather than just as a debugging tool. The re-player mimics the behavior of the application across a variety of storage systems by mapping multiple processes to multiple threads running on a single node. We show average replay error for parallel applications is below 30%."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Master of Science"]},{"key":"dc:title","label":"Title","values":["MPIOR: A Framework to Analyze File System Performance of MPI Applications"]}]}],"canonical_facts":{"dc:contributor.committeechair":["Varadarajan, Srinidhi"],"dc:contributor.committeemember":["Tilevich, Eli","Ribbens, Calvin J."],"dc:contributor.department":["Computer Science and Applications"],"dc:creator":["Banerjee, Shankha"],"dc:date.accessioned":["2014-03-14T20:32:40Z"],"dc:date.available":["2014-03-14T20:32:40Z","2012-04-11"],"dc:date.issued":["2012-02-17"],"dc:description.abstract":["MPI I/O replay (MPIOR) is an I/O performance modeling and prediction tool used to trace and replay a parallel application to determine application performance under a new I/O sub system. The trace collector deduces synchronization inter-dependencies between nodes and I/O demands placed by each node on the storage subsystem. It uses a novel runtime graph traversal technique to filter and log only those MPI calls that affect I/O, thus substantially reducing both the number of runs and the size of the trace file. Unlike other such tools, MPIOR collects a valid trace in a single run and it does not rely on node sampling or I/O sampling. MPIOR's post processing engine analyzes the trace files and sets up the re-player. Due to minimal overhead for trace collection, MPIOR can be used during production runs rather than just as a debugging tool. The re-player mimics the behavior of the application across a variety of storage systems by mapping multiple processes to multiple threads running on a single node. We show average replay error for parallel applications is below 30%."],"dc:description.degree":["Master of Science"],"dc:identifier.other":["etd-03162012-020243"],"dc:identifier.uri":["http://hdl.handle.net/10919/31484"],"dc:publisher":["Virginia Tech"],"dc:rights":["In Copyright"],"dc:rights.uri":["http://rightsstatements.org/vocab/InC/1.0/"],"dc:subject":["I/O","replay","MPI","trace"],"dc:title":["MPIOR: A Framework to Analyze File System Performance of MPI Applications"],"dc:type":["Thesis"],"thesis:degree_discipline":["Computer Science and Applications"],"thesis:degree_level":["masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Virginia Polytechnic Institute and State University"]},"updated_at":"2026-07-22T22:18:52Z"}